France: New study sheds light on membership of employers’ organisations

Reliable membership figures for employers’ organisations are vital in order to determine representativeness. A recently published analysis by Dares of data from a 2011 survey revealed that 44% of companies in France are affiliated to at least one employers’ organisation. The study found important differences between employers’ organisations in terms of coverage, sector representation and member company size.

Introduction

In September 2015, the Ministry of Labour's statistical directorate (Dares) published a detailed analysis of data from a 2011 survey on the membership of employers' organisations (in French, 928 KB PDF). 

Employers' organisations

There are three main employers’ organisations at national level in France:

• Movement of French Enterprises (MEDEF);
• Confederation of Small and Medium-sized Enterprises (CGPME);
• Craftwork Employers’ Association (UPA)

Other important actors include the National Union of Liberal Professions (UNAPL) and the Union of Employers of the Social Economy (UDES, formerly USGERES) representing employers from the non-profit sector. These organisations are national confederations made up of a large number of sectoral and/or regional groups of employers. Individual companies cannot directly join a national confederation, making it difficult to keep a record of total membership. In addition, a number of companies are affiliated to more than one organisation that might, in turn, be affiliated to competing national organisations. The complex structure of French employers’ organisations is one of the main reasons for the scarcity of reliable and comprehensive data on their membership.

Representativeness

Recent changes in French industrial relations legislation have introduced the principle that the representativeness of social partner organisations is subject to quantitative indicators. Since 2008, in order for a trade union to be representative and participate in company-level bargaining, it must obtain at least 10% of the votes in workplace elections. This threshold is set at 8% for bargaining at sectoral and national levels. As of 2017, employers’ organisations will have to organise at least 8% of all companies in a given sector to be considered representative. Hence, reliable membership figures will no longer be  merely of statistical interest, as correct information on how many companies a particular employers' organisation actually represents will become an issue of great political importance for social dialogue.

Key findings

Membership levels

The survey found that, in 2011, 44% of all private sector companies with more than 10 employees and employing 56% of the total French workforce were a member of an employers’ organisation. MEDEF had the highest membership figures, representing some 26% of all companies and 40% of all workers, followed by CGPME with 17% and 18% respectively (Figure 1).The survey also found that roughly half of MEDEF’s members and three-quarters of CGPME's affiliates in 2011 were associated with at least one other confederation. The specialised confederations for liberal professions and the non-profit sector had higher rates of exclusive membership The extent of multiple affiliations revealed by the Dares survey suggests that previous estimates, which added the total membership figures of all confederations, had resulted in significant overestimates of the representativeness of French employers’ organisations.

Figure 1: Coverage rate of French employers’ organisations, 2011

Note: The ‘non-profit’ category includes UDES and two smaller confederations.

Source: DARES

These figures were drawn from the management questionnaire of the REPONSE survey, which is similar to the British Workplace Employment Relations Study (WERS). The main advantage of using survey data is that it allows a bottom–up approach in which companies report their affiliation to one or several employers’ organisations. In most cases, density figures are taken from self-reported membership statistics. This raises concerns about reliability and, in particular in the French case, double counting of membership (see below). Moreover, the complex structure of the national confederations makes it unlikely that the top-level associations have reliable information on the number of their affiliated companies.

Membership distribution

Table 1 (below) shows the distribution of membership of employers’ organisations with regard to a range of company characteristics.

Size

Somewhat surprisingly, MEDEF records a stronger than expected membership among small and medium enterprises (SMEs), with 50% of its members employing between 11 and 49 staff. MEDEF is generally  perceived as an organisation that represents primarily the interests of large companies, a perception which may be due to the composition of its leadership (see below). According to official statistics, 83% of French companies have 10–49 employees (to increase comparability, companies with fewer than 10 employees were excluded from the calculations). Moreover, the Dares study found that many of the SMEs affiliated to MEDEF were also members of CGPME. Double membership was particularly common in the metal sector, in which even the most important employers' federation, the Union of Metallurgy and Mining Industries (UIMM) was affiliated to both MEDEF and CGPME. Exclusive MEDEF membership was found to be most prevalent among large companies or workplaces belonging to large groups.

In 2011, CGPME and UPA recruited about two-thirds of their members from the same company size brackets, in line with their political priorities which focus on small enterprises. When all employers’ organisations are taken into account, there is a slight representation bias towards larger companies, although it is weaker than might be expected (Figure 2).

Figure 2: Coverage rate of French employers’ organisations by size of company

Notes: Size brackets differ slightly. The category ‘fewer than 49 employees’ covers companies with 11 or more employees for the density figures and 10 or more workers for data on company structure.

Source: DARES and INSEE, author's own calculations

Sector

Both MEDEF and CGPME had a relatively broad membership base in 2011, with strongholds in the industrial, wholesale and retail, and construction sectors (Table 1). UPA had a relatively large share of its members in the transport sector but was weak in the industrial sector.

Employers’ organisations in the non-profit sector provide a range of services to their members, in particular in the area of quality control through the concession of labels and certificates. The three non-profit organisations were particularly strong in 2011 in the healthcare sector, which explains the high concentration of membership in education, health and social activities. 

Density

The density of employers' organisation membership was generally low in financial services in 2011 (Table 1). The general service sector was characterised by small employers’ organisations that were not affiliated to any of the large national confederations.

Age of company

A striking feature of membership patterns across all the employers' organisations featured in Table 1 is that membership was relatively low in 2011 in companies that had been established relatively recently (that is, within the past 10 years). Companies that had been established for 10–49 years formed the largest group in all the organisations (Table 1).

Industrial relations

An often cited argument says that deciding to join an employers’ organisation may be in response to the formation of organised labour institutions within the company. Members of MEDEF recorded high levels of trade union presence (43%) in 2011 compared with those in the total economy (31% according to an earlier Dares study based on the same data (in French, 1.13 MB PDF)). The even higher trade union presence reported in the non-profit sector is probably less surprising given the particular character of non-profit organisations and an assumed high number of cooperatives.

The Dares survey found that roughly half of the companies that had joined an employers’ organisation were a family business (51%); the proportion was particularly high among members of UPA, CGPME and MEDEF (66%, 65% and 56%, respectively) (data not shown in Table 1).

Source: DARES

Perception of low representativeness

Representatives of almost three-quarters (73%) of the companies surveyed by Dares stated that they perceived the representativeness of employers’ organisations as 'weak' or 'very weak'. Affiliation slightly increased the proportion of replies that believed their organisation represented their sector in a 'strong' (21%) or 'very strong' (25%) way compared with 17% on average.

This finding is in line with a recent analysis of the leadership of MEDEF and its affiliates (in French), whichound that the organisation's executives are predominantly male (98%, compared with 85.5% of leaders of French companies), relatively old (58.5 years old compared with 47 years old compared with all companies), graduates from prestigious universities and former leading managers of large groups with more than 500 staff. The study's authors concluded that MEDEF’s leadership might be representative of France’s large public companies but not of its SMEs.

Engagement in tripartite institutions

Research for a forthcoming Eurofound report has found that a relatively high proportion of union members in France are union activists (roughly one in five). For employers, the Dares survey found that some 29% of company representatives from affiliated businesses hold at least one mandate in an external organisation (compared with 21% for all businesses). Examples of such posts include a position in a chamber of commerce (8% for affiliated businesses compared with 5% in all businesses), on the supervisory board of another company (1% and 13%) or in a tripartite institution (13% and 8%).